Posted	Nick	Remark
#openstack-nova - 2019-03-15
01:01:20	takashin	gmann: Thanks. I will fix it.
01:02:39	openstackgerrit	Takashi NATSUME proposed openstack/nova master: Add description about sort order in API ref guideline https://review.openstack.org/627282
01:31:01	openstackgerrit	Fan Zhang proposed openstack/nova master: Retry after hitting libvirt error VIR_ERR_OPERATION_INVALID in live migration. https://review.openstack.org/612272
01:35:52	openstackgerrit	Julia Kreger proposed openstack/nova master: [ironic] ignore conflict upon cleaning up node https://review.openstack.org/643484
01:40:45	openstackgerrit	zhaixiaojun proposed openstack/nova master: Fix bug: non-admin user can not filter instance by instance-uuid https://review.openstack.org/642403
01:51:50	openstackgerrit	zhaixiaojun proposed openstack/nova master: Fix bug: non-admin user can not filter instance by instance-uuid https://review.openstack.org/642403
02:45:14	openstackgerrit	zhufl proposed openstack/nova master: Pass kwargs to exception to get better format of error message https://review.openstack.org/643501
02:52:51	openstackgerrit	Merged openstack/nova stable/queens: Fix WeighedHost logging regression https://review.openstack.org/641359
04:13:24	openstackgerrit	Merged openstack/nova master: Add functional test to delete a server while in VERIFY_RESIZE https://review.openstack.org/615644
05:09:56	openstackgerrit	Merged openstack/nova stable/pike: Migrate nova v2.0 legacy job to zuulv3 https://review.openstack.org/620579
07:17:45	openstackgerrit	zhaixiaojun proposed openstack/nova master: Fix bug: non-admin user can not filter instance by instance-uuid https://review.openstack.org/642403
08:01:18	kashyap	efried: dansmith: Caught up with the scrollback on #openstack-meeting. And this is the Intel's equivalent for AMD's SEV: "MKTME": https://www.redhat.com/archives/libvir-list/2019-February/msg01730.html
08:01:59	kashyap	dansmith: efried: The enablement of "MKTME" (Multikey Total Memory Encryption) is not merged in kernel itself yet.
08:02:36	kashyap	Then comes the design and implementation part in QEMU and libvirt (which is what the above libvirt mailing list URL points to).
08:17:39	openstackgerrit	Yongli He proposed openstack/nova master: Clean up orphan instances https://review.openstack.org/627765
08:45:18	bauzas	good morning Nova
08:47:52	hrw	morning
10:07:45	kashyap	Probably it's OK to not check too. "Distraction is the bane" ;-)
10:18:38	hrw	;D
10:18:48	hrw	zuul -> irc gate would be lovely
11:11:28	openstackgerrit	Matt Riedemann proposed openstack/nova master: Fix ProviderUsageBaseTestCase._run_periodics for multi-cell https://review.openstack.org/641179
11:11:29	openstackgerrit	Matt Riedemann proposed openstack/nova master: Improve CinderFixtureNewAttachFlow https://review.openstack.org/639382
11:11:30	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add functional recreate test for bug 1818914 https://review.openstack.org/641521
11:11:30	openstack	bug 1818914 in OpenStack Compute (nova) "Hypervisor resource usage on source still shows old flavor usage after resize confirm until update_available_resource periodic runs" [Low,In progress] https://launchpad.net/bugs/1818914 - Assigned to Matt Riedemann (mriedem)
11:11:30	openstackgerrit	Matt Riedemann proposed openstack/nova master: Remove unused context parameter from RT._get_instance_type https://review.openstack.org/641792
11:11:30	openstackgerrit	Matt Riedemann proposed openstack/nova master: Update usage in RT.drop_move_claim during confirm resize https://review.openstack.org/641806
11:11:31	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add Migration.cross_cell_move and get_by_uuid https://review.openstack.org/614012
11:11:31	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add InstanceAction/Event create() method https://review.openstack.org/614036
11:11:32	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add Instance.hidden field https://review.openstack.org/631123
11:11:32	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add TargetDBSetupTask https://review.openstack.org/627892
11:11:33	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add CrossCellMigrationTask https://review.openstack.org/631581
11:11:33	openstackgerrit	Matt Riedemann proposed openstack/nova master: Execute TargetDBSetupTask https://review.openstack.org/633853
11:11:34	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add can_connect_volume() compute driver method https://review.openstack.org/621313
11:11:34	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_dest compute method https://review.openstack.org/633293
11:11:35	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add PrepResizeAtDestTask https://review.openstack.org/627890
11:11:35	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add prep_snapshot_based_resize_at_source compute method https://review.openstack.org/634832
11:11:36	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add nova.compute.utils.delete_image https://review.openstack.org/637605
11:11:36	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add PrepResizeAtSourceTask https://review.openstack.org/627891
11:11:37	openstackgerrit	Matt Riedemann proposed openstack/nova master: Refactor ComputeManager.remove_volume_connection https://review.openstack.org/642183
11:11:48	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add revert_snapshot_based_resize conductor RPC method https://review.openstack.org/638047
11:11:48	openstackgerrit	Matt Riedemann proposed openstack/nova master: Revert cross-cell resize from the API https://review.openstack.org/638048
11:11:49	openstackgerrit	Matt Riedemann proposed openstack/nova master: Confirm cross-cell resize while deleting a server https://review.openstack.org/638268
11:11:49	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add CrossCellWeigher https://review.openstack.org/614353
11:11:50	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add cross-cell resize policy rule and enable in API https://review.openstack.org/638269
11:12:16	mriedem	gibi: bauzas: alex_xu: the bottom 4 changes in the cross-cell-resize series are test-only changes and are generally useful (not directly related to cross-cell but i do need them in that series later) ^
11:12:31	mriedem	it would help to land some of what we can while we can since that series is currently at 44 changes
11:13:11	mriedem	dansmith: from when you awaken from your cryochamber ^
11:13:15	mriedem	*for when
11:20:36	kashyap	The re-proposed SEV spec needs some further re-design (to take into account a blocker raised by DanPB): https://review.openstack.org/#/c/641994
11:35:47	stephenfin	mriedem: RE: https://bugs.launchpad.net/nova/+bug/1819794, I assume there's still an open to figure out why https://bugs.launchpad.net/nova/+bug/1819794/comments/2 is happening?
11:35:47	openstack	Launchpad bug 1819794 in OpenStack Compute (nova) "nova-next job fail on Ubuntu Bionic" [Medium,Confirmed]
11:36:02	mriedem	stephenfin: correct, it's not fixed
11:36:36	stephenfin	mriedem: Okidok. Won't get to that today but I'll try take a look next week
11:44:35	efried	kashyap: ack, thanks for the info. dansmith, does the dev stage of that project affect your opinion on landing the SEV work?
11:45:50	efried	are you thinking we should perhaps call it something more generic, so that when Intel's answer arrives on the scene, we can implement it under the covers without changing the UI at all?
11:45:51	kashyap	stephenfin: I'll look at that Bionic thingie; there are two things there (VMX thing and the misconfig of TLS certs)
11:46:08	stephenfin	kashyap: It's all yours
11:46:27	kashyap	stephenfin: Nod, first clearing out existing queue of stuff
11:47:03	kashyap	efried: On the "calling it more generic", that's what I wanted to plant a flag about -- for us (Nova) to consider _if_ we can work out a generic mechanism
11:48:51	kashyap	efried: From a quick chat with DanPB yesterday, he said Intel "MKTME" takes quite a different approach to AMD's SEV -- and what we're going to get in common in the end is up in the air.
11:49:19	efried	kashyap: Well, yeah, I would think the outward-facing bits, like the compute capability trait?, would be called <generic>, and then under the covers we
11:49:20	efried	if <generic>:
11:49:20	efried	if AMD:
11:49:20	efried	turn on SEV
11:49:20	efried	elif Intel:
11:49:20	efried	turn on MKTME
11:49:24	efried	oh
11:49:28	efried	well then
11:52:47	kashyap	efried: On the outward-facing bits like the capability trait (although I don't know an awful lot about it) -- yes, it is worth at least calling out in the spec
11:53:37	kashyap	... that shows we've considered Intel's equivalent, with whatever limited info we have now.
11:53:56	kashyap	efried: Might want to add your above pseudo-code comment in the spec
11:54:15	efried	kashyap: I suppose I'll have to finally *read* the spec :P
11:54:59	kashyap	efried: Haha. I'll admit, I was not paying attention to it at all (as my plate is already overflowing with stuff).
11:55:16	kashyap	efried: I slowly got into it, as I was pinged about handling machine types, and then I noticed a few things
11:56:02	cdent	wonders how kashyap is managing to have just one plate
11:56:08	cdent	I will study
11:56:12	efried	kashyap: I'm very new to "caring about libvirt at all", and it seemed to be a fairly simple followon to... some other spec I can't remember where we did a capability trait thing.
11:56:16	kashyap	And then I saw things (like adding the wrapper around libvirt's getDomainCapabilities() API) could be split out -- as it can be used for other libvirt driver work
11:56:54	kashyap	cdent: Hehe, I split my time b/n Nova (Gerrit-based patch workflow) and low-level virt (mailing-list based patch workflow)
11:57:11	cdent	sounds very much like two plates
11:57:41	kashyap	Yeah, a bit brain-warping. The joy of applying 40 patches from an e-mail thread :-)
11:57:48	efried	but yeah, with the plate full thing, stuff involving the bowels of libvirt tends to be the first to slide off the edge.
11:57:56	kashyap	But ... if you're using tools like `mutt`, et al. That makes it blazing fast, though
11:58:54	kashyap	efried: Can understand on the "caring about libvirt" thing. The "being simple to follow" depends on what one is looking at, though ;-)
11:59:34	efried	kashyap: Well, yes, simple from a "how we used the placement-isms" perspective, mainly because I've been right in the middle of that business for a whiile.
12:01:13	kashyap	Ah, gotcha; in that sense.
12:02:18	openstackgerrit	Matt Riedemann proposed openstack/nova master: docs: cleanup driver parity scope section https://review.openstack.org/643566
12:02:53	kashyap	efried: And thanks for reviewing the cache mode change series.
12:03:18	efried	kashyap: Sure thing (speaking of bowels of libvirt)
12:03:38	kashyap	:-)
12:04:18	kashyap	efried: Warning: That SEV spec has _tons_ of hypervisor-related concepts. So expect to sink some 2 hours or more, if you're reading with full concentration
12:04:37	efried	this is probably why I never finished it the first time around.
12:04:48	efried	I was like, "whoah, this stuff is over my head; I hope kashyap reviews it."
12:04:51	kashyap	(And as we speak, there is potentially one more thing to be fixed in libvirt/QEMU; libvirt folks are testing)
12:05:25	openstackgerrit	Matt Riedemann proposed openstack/nova master: Remove resize caveat from conductor docs https://review.openstack.org/643567
12:05:44	kashyap	efried: I've commented on parts of it; for now you can ignore and tackle other high-prio stuff.
12:06:00	efried	thanks
12:06:01	kashyap	I still need to review all the firmware / OVMF and other memory requirements stuff.
12:06:07	kashyap	Will get to it next week
12:09:29	stephenfin	mriedem: wrt all things docs, there's a nice albeit lengthy of all things metadata here that could benefit from your attention https://review.openstack.org/#/c/640730/

Earlier   Later